Abstract
BACKGROUND: Some studies have detected associations between in utero antiretroviral therapy (ARV) exposure and birth defects but evidence is inconclusive. METHODS: A total of 2202 human immunodeficiency virus (HIV)-exposed children enrolled in the Pediatric AIDS Clinical Trials Group 219 and 219 C protocols before 1 year of age were included. Birth defects were classified using the Metropolitan Atlanta Congenital Defects Program coding. Logistic regression models were used to evaluate associations between first trimester in utero ARV exposure and birth defects. RESULTS: A total of 117 live-born children had birth defects for a prevalence of 5.3% (95% confidence interval [CI]: 4.4, 6.3). Prevalence did not differ by HIV infection status or overall ARV exposure; rates were 4.8% (95% CI: 3.7, 6.1) and 5.8% (95% CI: 4.2, 7.8) in children without and with first trimester ARV exposure, respectively. The defect rate was higher among children with first trimester efavirenz exposure (5/32, 15.6%) versus children without first trimester efavirenz exposure (adjusted odds ratio [aOR] = 4.31 [95% CI: 1.56, 11.86]). Protective effects of first trimester zidovudine exposure on musculoskeletal defects were detected (aOR = 0.24 [95% CI: 0.08, 0.69]), while a higher risk of heart defects was found (aOR = 2.04 [95% CI: 1.03, 4.05]). CONCLUSIONS: The prevalence of birth defects was higher in this cohort of HIV-exposed children than in other pediatric cohorts. There was no association with overall ARV exposure, but there were some associations with specific agents, including efavirenz. Additional studies are needed to rule out confounding and to evaluate newer ARV agents.